PROJECT SUMMARY/ABSTRACT Esophageal adenocarcinoma (EAC) is a lethal cancer with poor outcomes (5 year survival <20%), when diagnosed after the onset of symptoms, but survival is excellent when diagnosed early. Intestinal metaplasia, or Barrett?s esophagus (BE), is the only known precursor of EAC, and progresses to EAC via development of dysplasia. EAC can be prevented by endoscopic therapy of dysplasia. Hence endoscopic screening for BE and endoscopic surveillance to detect dysplasia and EAC are recommended. However, > 60% of prevalent BE remains undiagnosed and 90% of all EAC cases are diagnosed outside a BE surveillance program. The major barrier to BE screening is the invasiveness and high cost of endoscopy. Further, screening is recommended only in those with chronic gastroesophageal reflux (GERD), despite 50% of BE/EAC patients not reporting GERD symptoms. Endoscopic surveillance misses 33% of prevalent EAC & dysplasia, due to the patchy distribution of dysplasia/EAC, and inadequate biopsy sampling. Hence the overall effectiveness of endoscopic surveillance is also severely compromised. We used reduced representation bisulfite sequencing (RRBS) to identify a panel of methylated DNA markers (MDMs) of BE and dysplasia/EAC followed by validation. MDM panels were highly discriminant (AUCs > 0.9) for BE and prevalent dysplasia/EAC. When assayed on esophageal cytology specimens obtained via a sponge on string (SOS) device, BE was detected with high accuracy (AUCs 0.97-1.0) in two case control studies done in referral populations. The FDA approved SOS device (Capnostics, Doylestown, PA) is a 25 mm polyurethane sponge compressed in a dissolvable capsule shell, which expands into a sphere in the stomach after being swallowed. When pulled out through the mouth via an attached string, sampling of the entire esophageal mucosa is achieved. The nurse-administered SOS test is safe and well tolerated with high participation rates (65%). Hence our central hypothesis is that novel discriminant MDMs assayed on esophageal cytology specimens obtained via the SOS device will enable accurate BE and dysplasia/EAC detection, in a screening population with and without chronic GERD. We will test this hypothesis by three specific aims. In specific Aim 1, we will measure the positive and negative predictive value of the SOS BE test in a screening eligible population from primary care clinics in the Mayo Health System and compare these values in those with and without GERD. In specific Aim 2 we will identify clinical and demographic factors, particularly GERD, influencing the accuracy a predetermined SOS BE test prediction algorithm. In Specific Aim 3, we will measure the accuracy of MDMs for the detection of dysplasia/EAC in BE, using the SOS device. Utilizing an innovative, minimally invasive (non-endoscopic) and molecular approach, this proposal will favorably impact BE detection and surveillance, enabling effective treatment, and improved EAC outcomes.